Skip to Content

Hope For Nonaddictive Opioid Painkillers

No one needs to be told about the opioid painkiller problem in this country. There are legal, commerical, regulatory, and ethical ways to look at it, but from a pharmacological standpoint, the whole thing would be a lot easier to deal with if there were any highly effective non-addictive painkillers. But that’s exactly what we lack. Pain, as a therapeutic area, has been very painful indeed over the years, with a lot of good ideas that have wiped out in actual clinical trials. There are actually several different kinds of pain signals that differ in both degree and kind, and we still have not untangled what’s going on with all of them.

You could imagine two general approaches to the problem. Searching for new pain mechanisms is the one that most people have been trying, in the hopes that these will not have the addiction potential that the opioid receptors do, but that’s what’s led to all sorts of ion-channel (and other receptor) failures. Another way to approach the problem would be to find a way to hit those opioid receptors without causing addiction, but that’s not easy to picture. Last year, though, there was a report of a compound (an orvinol derivative) that simultaneously hits mu-opioid receptors and nociceptin receptors, and in primate models it showed analgesia but not addictive potential, which is definitely interesting.

This new paper in Science takes another ingenious approach (here’s a story on it at Ars Technica). The authors, a team from Berlin, noted that inflamed and injured tissues generally have lower (more acidic) pH than normal. They hypothesized that the mu-opioid receptor itself might have different behavior under those conditions, which might be exploited by a compound that would be less active in normal tissues. A fluorinated derivative of fentanyl seemed to meet those conditions in cell assays, showing pH-sensitive activity. When dosed in rodents, it seems to have a very different profile from fentanyl itself, showing pain relief only at the site of injury. A non-centrally-acting opioid antagonist can only partially reverse the effects of fentanyl, since some of those are from inside the CNS, but the same agonist seems to completely reverse the effects of the fluorinated compound. And it also shows no addiction potential, GI or respiratory effects, or sedation, all of which are hallmarks of classic opioid agonists. (The compound from the 2016 paper mentioned above was also devoid of these effects).

These sorts of results are very hopeful, and I hope that such compounds advance to human trials as soon as possible. Pain is an awful problem, but by treating it with addictive compounds medicine has added another hideous problem on top of it. If we can get to the point of being able to regard classical opioid drugs as barbaric remnants of an earlier age, that would be a great improvement.

Update: be sure to have a look at the comments. There are some good arguments for why these approaches may not translate, so this will be worth watching to see how they progress.

61 comments on “Hope For Nonaddictive Opioid Painkillers”

  1. Anon says:

    The only 100% effective, 100% non-addictive painkiller I’m aware of, is death — which has the rather undesirable side effect of not living.

    1. PD says:

      Infamous quote from inside our shop that came up during a candidate advancement meeting: “Death is not a side effect.”

  2. SirWired says:

    Wow. I’m no pharmacologist (or even a chemist, or even a janitor at a drug lab), but this sounds very promising. Previous approaches to “non-addictive” opioids seem to have been based on either wishful thinking (Heroin started out being marketed this way, way back in the early 1900’s) or forlorn hopes, by trying to prevent addiction by trying to make abuse unpleasant/difficult, vastly underestimating the resourcefulness of addicts, and doing nothing to address the addictive potential of opioids as a class.

    I’m sure the creative or desparate will still find a way to abuse these (there are people that satisfy opioid cravings by ingesting crazy-high amounts of loperamide), but it should go a long way (if approved) to prevent “accidental” addictions following an Rx after surgery or an accident.

    That said, if this still has similar problems with the need for ever-increasing dosages to hold pain relief constant? Even if it’s not addictive to the brain, and won’t kill you in an OD, it may not solve the problem of tolerance to the medication for actual pain relief.

    1. tangent says:

      I ran across this paper, and now I’m pretty surprised that Imodium hasn’t yet been abused to hell and gone: dosing with quinidine inhibits a BBB efflux transporter and creates central activity for loperamide. (Respiratory depression is the paper’s measure, but I can’t see why this wouldn’t have the full effect spectrum.)

      1. Thomas says:

        Actually there some reports of abuse of Loperamide (800mg/day!!), but not taken with a Pg-P inhibitor. (BMJ Case Rep 2015. doi:10.1136/bcr-2015-209705)
        I think one of the trigger for dependency are the Tmax and Cmax (or the sharp rise of C vs time), I guess with loperamide+a Pgp these would be blunted?

      2. Samuel71 says:

        This is well-known among opioid users, but is not terribly efficacious.

  3. Mister B. says:

    Naive question, but how good are the animal models in pain treatment ?
    (Compared to the AD ones, for example)

    Thank you

    1. Derek Lowe says:

      No, that’s not at all a bad question. That’s one of the issues that seems to have led to failures in the past, but the hope is that since these are still mu-opioid-acting drugs that they’ll translate well, as opposed to some of the completely different mechanisms.

      1. Mol Biologist says:

        IMO CCR9 and μ-opioid receptors are very similar targets. So you can’t escape addiction since they are covered by same biological mechanism.

    2. ex-pain scientist says:

      Generally, animal models for pain do not translate to human very well. That’s been a problem for the industry and it’s one of the reasons the pain therapeutic sector has experienced “pain” in recent years. Even though pain symptoms are acute, the underlying pharmacology can be elusive and complex, depending on causation as well as chronic pain states.

      1. Mister B. says:

        Thank you both for your answer !

        Do you have any good references for an introduction to pain-related biology and associated drugs ?

        1. Istvan Ujvary says:

          An excellent 2001-overview on animal models of nociception:

      2. anon says:

        It is generally true that pain models are unreliable,

        Activity of MU OPIOID AGONISTS generally track well, though. Human and rodent receptors are highly homologous and have similar distribution and expression.

  4. bhip says:

    Interesting but I have my doubts. When I moved therapy areas from immunology/inflammation to pain/depression, I insisted that the animal models in pain couldn’t be significantly more useless than the immunology models (hello, EAE & CIA- I’m talking to you!!)- I was wrong. The translational value of pain models to clinical efficacy is really bad. There are other peripheralized opioid agonists (i.e. loperamide= Imodium; full agonist at the mu opioid receptor with very limited CNS access; pegylated agonists from Nektar) showing good preclinical activitity. Pain relief via mu opioid agonists seems to be primarily (only?) centrally mediated.

    1. Mike says:

      I had the same thought. Loperamide is a mu-opioid receptor agonist that has poor central CNS activity (I’ve heard that this is due to very efficient efflux pumps at the blood brain barrier). It does, however, have strong peripheral activity which is why it constipates you so effectively (mu-opioid receptors in the gut).

      However, it’s not known to have any analgesic activity. This is strong evidence that this paper’s findings won’t translate well into humans.

      1. bhip says:

        Hi Mike-preclinically, loperamide shows some analgesic effects when given intravenously or locally. Loperamide has poor oral bio-availability- it acts locally on mu opioid receptors in the gut.

        1. NJ Micro Guy says:

          Well if you drink enough of it you can still get a sufficient quantity through the BBB to get high:

          1. myma says:

            50 to 300 pills per day to get high?
            … 300 pills a day of immodium?

            and only 2 stops me up just fine. I can’t even imagine that many tablets in my stomach at the same time , what kind of volume – – that’s like a big cereal bowl full of tablets to get high? wow.

        2. tangent says:

          Wow, have to update my mental model!

          Looks like it’s strongly effective intrathecally. In some papers, I see suggestion that it may not actually be acting as a mu ligand there — contradictory reports of whether it’s affected by naloxone.

          (Yeah, I was writing this line of work off as bizarre and hopeless given (the assumption) that opioids are exclusively centrally-acting. Should have thought maybe they did talk a pharmacologist better than me…)

  5. SpaceCore says:

    Having a fentanyl addict member of my family who, to make the short story : became addict to morphine over the years because of migraines (wich aren’t a simple problem either), did rehab, had a cancer and painkillers prescribed again (side note: we live in France, not related to the opioid crisis)- i really hope these studies could finally give something. Being myself a biology-interested chemistry student I look forward to these compounds !

    Anyway, It’s been a few years and the health problems are gone apart from that so it’s still nice we have beaten cancer. Also, great blog, discovered it reading all the “Things I Won’t Work With” articles and found them very fun readings ! I also thoroughly enjoy the “Cram more nitrogen in that” kind of articles and the ones about unusual compound like drugs with phosphorus or metals in them. Keep going !

    1. Mister B. says:

      Hi french fellow 😉
      Glad to see some compatriots reading this excellent blog !

  6. Spacenaut says:

    Any hope in the Nav1.7 space for a solution?

    1. HTSguy says:

      NaV1.7 blockers have not worked as well in clinical trials as was initially expected. One possible reason is that the mechanism is more complicated than originally thought and may require nearly complete blockage (as is seen with mutations causing CIP).

    2. ex-pain scientist says:

      The clinical path of Nav1.7-specific inhibitors seems to be complicated, given that it’s taken years for anyone in development to make it beyond PhaseII. Also, word on the street says that a large biopharmaceutical company was unable to corroborate the Nav1.7 mab results of the Lee group at Duke. While Nav1.7 is likely a key part of the nociception equation, it’s only partial – it will likely be in combination with a complementary pain target (that may include regulation of pain signalling in the CNS). Not sure if the experts have changed their tune about Nav1.7 pharmacology again:

  7. Just wondering if there’s a connection with the recent announcement from Heptares on the topic of new approaches to discovering pain therapeutics:

    1. Mol Biologist says:

      I guess you are right. Let’s say it is a farewell party for μ-opioid receptor story.

  8. SBDD Rob says:

    Here is a very interesting paper that identifies a novel non-alkaloid lead scaffold that exhibits an analgesic profile and is devoid of respiratory depression and addictive side effects. New and more chemically tractable scaffolds with the desired pharmacology certainly have the potential to become breakthrough therapeutics.

    Structure-based discovery of opioid analgesics with reduced side effects

    Aashish Manglik, Henry Lin, Dipendra K. Aryal, John D. McCorvy, Daniela Dengler, Gregory Corder, Anat Levit, Ralf C. Kling, Viachaslau Bernat, Harald Hübner, Xi-Ping Huang, Maria F. Sassano, Patrick M. Giguère, Stefan Löber, Da Duan, Grégory Scherrer, Brian K. Kobilka, Peter Gmeiner, Bryan L. Roth & Brian K. Shoichet

    Nature 537, 185–190 (08 September 2016) doi:10.1038/nature19112

    Morphine is an alkaloid from the opium poppy used to treat pain. The potentially lethal side effects of morphine and related opioids—which include fatal respiratory depression—are thought to be mediated by μ-opioid-receptor (μOR) signalling through the β-arrestin pathway or by actions at other receptors. Conversely, G-protein μOR signalling is thought to confer analgesia. Here we computationally dock over 3 million molecules against the μOR structure and identify new scaffolds unrelated to known opioids. Structure-based optimization yields PZM21—a potent Gi activator with exceptional selectivity for μOR and minimal β-arrestin-2 recruitment. Unlike morphine, PZM21 is more efficacious for the affective component of analgesia versus the reflexive component and is devoid of both respiratory depression and morphine-like reinforcing activity in mice at equi-analgesic doses. PZM21 thus serves as both a probe to disentangle μOR signalling and a therapeutic lead that is devoid of many of the side effects of current opioids.

    1. tangent says:

      I’m not in the field, is there previous reason to think that the side effects are off-target effects, due to the beta-arrestin signaling rather than to mu activation itself? If these folks have a more selective ligand they could can now discover that, certainly conceivable, but it sounds like they’re saying this was previously known or suspected.

      Of course there’s a long history of underestimating the abuse liability of a new drug, so we’ll see how that goes. Just a layperson, but I will be seriously surprised if morphine’s central antinociceptive effects — reducing the emotional response to pain — are separable from its effects against other negative emotional responses.

      1. MN says:

        To your question, take a look at Oliceridine (TRV130) developed by Trevena, and that is currently in PhIII. There is a lot of expectation coming from that biased mu-opioid agonist approach or biased ligands at GPCRs in general.

        1. tangent says:

          Thanks, that does look pretty interesting. (I had no idea how “multiplex” the 5-HT2A receptor was either, per Wikipedia.)

          I wonder whether somebody has totally unofficially snorted it to get a real picture of how abusable it is. The animal models don’t have a great track record on that. I suppose it doesn’t much matter unless they’re trying to argue for a lowered schedule — ‘just’ reduced respiratory and GI side effects would be a huge benefit.

          1. tangent says:

            Figure 8 here for the respiratory depression:

        2. bhip says:

          Hi- TRV130 is not a biased agonist as claimed by Travena, it’s a partial agonist a la buprenorphine. Under questioning at a meeting at a recent meeting, they acknowledged as much.

          1. MN says:

            Interesting. Do you know the reasons of that backpedalling? Their original JPET paper seemed to show a full ago profile if I am correct. Is it related to their beta-arrestin recruitment assay or other observations?

          2. Emjeff says:

            You are incorrect; olceridine is indeed a biased ligand, and there are numerous papers out there showing the data.

          3. bhip says:

            They backpedaled (when pushed) because the dose response differentials for G-coupled response (~10 nM for cAMP) vs B-arrestin (~40nM) increase is not sufficiently different to support the claim. Its In vivo potency in pain vs depression of colonic motility are also equivalent (Fig 7; J Pharmacol Exp Ther 344:708–717). Travena has a vested interest in claiming it is biased as that is the basis of their company. The pharmacology they present is classic for a weak partial agonist. Nothing wrong with that- lots of drugs are partials including the aforementioned buprenorphine, morphine, albuterol, salmeterol, etc.

  9. David Watson says:

    Any word on NPFF-receptors lately? I’ve been working in technology for the last few years and missed out on quite a bit, apparently. I’m happy to see some progress!

  10. Kent G. Budge says:

    “The authors, a team from Berlin, noted that inflamed and injured tissues generally have lower (more acidic) pH than normal. They hypothesized that the mu-opioid receptor itself might have different behavior under those conditions, which might be exploited by a compound that would be less active in normal tissues”

    I know this is probably oversimplifying the biochemistry badly, but I now have this mental image of a bunch of stoned diabetics …

  11. Mikeb. says:

    Any progress on allosteric modulators of any of these pain targets? I thought the exciting idea wrt allosterism was the ability to unlock a treasure trove of new biology not possible with ant/ag -onists, such as greatly reducing the amount of an opiate needed for pain relief, thereby limiting addiction problems.

  12. McNutcase says:

    If, and it’s a big if, nonaddictive analgesics with potency similar to conventional opiates come along, then I, as a current user of opiates for chronic pain, will be throwing a party in celebration. I hate the current situation in which I need to constantly decide how much pain I’m willing to be in, how much social stigma I’m willing to deal with, and whether I can afford the undesirable side effects of conventional opiates. I’ve lost count of how many times I’ve spent days shaking, sweating, and feeling sick on the couch while I go cold turkey again. Knocking even one variable out of that pain calculus would immeasurably improve my quality of life.

    1. Mike Andrews says:

      My wife and I are right there with you. We both have Rheumatoid Arthritis, and she has Fibromyalgia and Lupus as bonuses. My pain can be pretty brutal, and hers is even worse, despite TNF inhibitors and many other disease-modifying anti-rheumatic drugs. Her pain control is pretty good much of the time, thanks to a non-addictive μ-agonist κ-antagonist. This Fentanyl derivative, if it gets to market, will be an absolute blessing for both of us, as she will have good pain management without injections, and I will have better pain control than ever before.

  13. Scott says:

    I suppose I should be thankful that I don’t have whatever the addictive factor is in my personality. I take Tramadol for chronic pain after I broke my back in a nasty fall (L1 burst fracture, left about half of L1 solid!), but I’m up at the safe limit to keep the pain under control 24/7.

    You *really* don’t want to know what I’d give to just not HURT all the [expletive deleted] time. That billion-dollar Powerball jackpot? Yeah, I’d give that up in a heartbeat to be pain-free.

  14. aairfccha says:

    The whole thing would be a lot easier to deal with if moralists wouldn’t get so hung up about drugs. Short of an acute overdose, opioids are fairly non-toxic and an overdose can be easily dealt with by an antagonist.

    Granted, this new version probably can’t be overdosed which is good, but most opioid problems (drug problems in general) can be described more accurately as prohibition problems and side effects of the pipe dream of a drug-free world.

    Heroin-assisted Treatment (HAT) a Decade Later: A Brief Update on Science and Politics
    “the discussed studies above have demonstrated in several different contexts that the implementation of HAT is feasible, effective, and safe as a therapeutic intervention. This should not be seen as a conclusion that could be taken for granted because many observers expected disastrous consequences from the provision of medical heroin prescription.”

      1. aairfccha says:

        Just skimmed it but yes, the term moralists includes the abstention priests of AA and NA.

  15. Anonymous says:

    Re: The Qs and Comments about animal models for pain. I say, bring back the good old days when you could do tail clamp, tail flick, hot plate, and other tests on your rodents! Drug discovery was so much easier then! The low-lying fruits — or rats — are gone and the real world of drug disco is more painful than it used to be. (My serious Q: do they still use those tail clamp, hot plate and other assays?)

  16. SedatedFMS says:

    How long before the woohoo halfwits pickup on the “nflamed and injured tissues generally have lower (more acidic) pH than normal.” line and start off the the alkalising your diet nonsense?

  17. Molecular says:

    What the the authors of the Science paper do not seem to realize (or at least do not discuss) is that monofluorination results in two stereogenic centers (the fluorinated carbon and the adjacent quaternary carbon). There is no mention of the stereochemistry of the compounds (except for occasional mention of ‘enantiomers’) or its influence on activity. This is certainly worth investigating further.

    1. Istvan Ujvary says:

      The authors of the Science paper are well aware of the existence of 4 stereoisomers. See the patent by Stein et al (Universitaetsmedizin Berlin) EP2559685

      1. Molecular says:

        They do indeed show al four stereoisomers in the patent, but as far as I could determine in my one-minute in-depth analysis of the patent they have no data on relative activities of the stereoisomers; this might be why they sort of swept this issue under the carpet in the Science paper…

  18. etc says:

    an upcoming NIH videocast that may be of interest (March 17):

    “America’s Opioid Overdose Epidemic: What We Know and How Health Practitioners Can Help”

    1. aairfccha says:

      My take on a first quick and dirty fix: Naloxone OTC, must-carry (and sell) for pharmacies, no questions asked.

      1. Vader says:


        Please tell me more about your species, and the planet it is native to. I see some fascinating differences with H. sapiens that are worth exploring.

        1. aairfccha says:

          Key difference: presence of a working brain.

  19. milkshaken says:

    I don’t believe this “pH explanation” is correct, and that sticking one F in the 3-position of piperidine ring, next to the propionanilide, has miraculous properties that suddenly make fentanyl compounds nonaddictive. From SAR it is known that adding cis-methyl susbstituent in the 3 position leads to more potent compounds, (one of the enantiomers of 3-cis-Me-fentanyl is about 10 times more potent than parent fentanyl)

  20. FarEasterner says:

    Cara Therapeutics ‘s local acting kappa opioid now in ph 3 sounds very promising.

  21. thegooddrugs says:


    I would love to see you do a piece on Kratom at some point in the future. Lots of miracle-touting around it, which throws up my BS meter immediately.

  22. metaphysician says:

    So. I am not a doctor or medical researcher, so bear with me if this is a stupid question.

    Given the evidence available, what is the probability that “non-addictive opioid painkiller” is actually a solvable problem at all? That is, do we have good reason to believe the analgesic properties are in any way separable from the addictive ones, in terms of the neurological pathways effected? Or conversely, do we have evidence to suggest that the dependency problems are “upstream” of where the opioids operate, and anything that effects pain in a similar way will produce addiction no matter how its formulated, because its the pain relief that is causing the addiction?

    ( I expect the answer is “Reply Hazy”, but I’m curious how much evidence there actually is on that fundamental question. )

    1. Barry says:

      the Shoichet paper in Nature claims that they have ligands that agonize the mu-Opiate receptor without signalling through beta-arrestin. If that holds up, it’s the first evidence I’ve seen that opiate-receptor-mediated analgesia is separable from addiction.

      1. Barry says:

        the Shoichet paper in Nature claims that they have ligands that agonize the mu-Opiate receptor without signalling through beta-arrestin. If that holds up, it’s the first evidence I’ve seen that opiate-receptor-mediated analgesia is separable from addiction.

        1. Mol Biologist says:

          There is only way to separate it from addiction if you restore brain energy metabolism toward elimination the deficiency or inflammation. No other way around.

Leave a Reply

Your email address will not be published. Required fields are marked *

Solve the math problem. * Time limit is exhausted. Please reload CAPTCHA.